Jump to contentJump to search

Software Contributions


ESP is the first general machine learning model for predicting enzyme-substrate pairs. The model achieves an accuracy of over 91% on diverse test data and outperforms previous models designed for specific enzyme families. To make predictions, ESP takes an enzyme amino acid sequence and the structure of a potential substrate as its input. The model can be applied either by using an easy accessible web server (esp.cs.hhu.de) or by using prediction functions implemented in Python (https://github.com/AlexanderKroll/ESP_prediction_function).


TurNuP is a machine learning model for predicting turnover numbers for enzymatic reactions. It is a general and organism-independent model that outperforms existing methods by accurately predicting turnover numbers for natural reactions of wild-type enzymes, even for enzymes dissimilar to those in the training set. TurNuP can be easily used via web server (turnup.cs.hhu.de).


PopGenome is an efficient Swiss army knife for population genetics data analysis, able to process individual loci, large sets of loci, or whole genomes. PopGenome not only implements a wide range of population genetics statistics, but also facilitates the easy implementation of new algorithms by other researchers. PopGenome is an R library optimized for speed (via the seamless integration of C code).
The software is available from the PopGenome homepage.


The R-package sybil is a Systems Biology Library for R, implementing algorithms for constraint based analyses of metabolic networks, such as flux-balance analysis (FBA) that is a mathematical technique to find fluxes in metabolic models at steady state.

Through fine-tuned communication with the optimization software, sybil is extremely fast when performing large sets of similar analyses. 
The software is available on CRAN. More information is available on the sybil homepage.


Many constraint based metabolic models are published in 'SBML' format ('*.xml'). Herewith is the ability to read, write, and check 'SBML' files in 'sybil' provided. More information is available on the sybil homepage.


Implementing cycle-Free flux balance analysis allows to efficiently remove thermodynamically infeasible loops from flux distributions.
sybilcycleFreeFlux is available on CRAN.


This extension implements different techniques of cost constrained flux balance analysis:
1. MetabOlic Modeling with ENzyme kineTics 'MOMENT' which uses enzyme kinetic data and enzyme molecular weights to constrain flux balance analysis(FBA).
2. An improvement of 'MOMENT' that considers multi-functional enzymes 'ccFBA'.
sybilDynFBA is available on CRAN.


The package sybilDynFBA is an extension to sybil in order to perform dynamic FBA.
sybilDynFBA is available on CRAN.


The R-package sybilFCF computes extreme pathways and coupled reaction sets.

TreeSnatcher Plus

TreeSnatcher Plus is a GUI-driven Java application that recognizes bifurcating and multifurcating trees in pixel images (like JPG) and outputs the tree in Newick format.
The software is available at the TreeSnatcher Plus homepage

Responsible for the content: